The present invention is in the field of battery technology, and more particularly in the area of materials for making high-energy electrodes for batteries, including metal-fluoride materials.
One type of battery consists of a negative electrode made primarily from lithium and a positive electrode made primarily from a compound containing a metal and fluorine. During discharge, lithium ions and electrons are generated from oxidation of the negative electrode while fluoride ions are produced from reduction of the positive electrode. The generated fluoride ions react with lithium ions near the positive electrode to produce a compound containing lithium and fluorine, which may deposit at the positive electrode surface.
Metal fluoride based batteries are an attractive energy storage technology because of their extremely high theoretical energy densities. For example, certain metal fluoride active materials can have theoretical energy densities greater than about 1600 Wh/kg or greater than about 7500 Wh/L. Further, metal fluorides have a relatively low raw material cost, for example less than about $10/kg. However, a number of technical challenges currently limit their widespread use and realization of their performance potential.
One challenge for certain metal fluoride materials is comparatively poor rate performance. Many metal fluoride active materials have electrochemical potentials greater than about 2.5 V because of their relatively large bandgap produced by the highly ionic bonding between the metal and fluorine, and in particular between a transition metal and fluorine. Unfortunately, one of the drawbacks to wide bandgap materials is the intrinsically low electronic conductivity that results from the wide bandgap. As a result of this low conductivity, discharge rates of less than 0.1 C are required in order to obtain full theoretical capacity. More typically, discharge rates of 0.05 C to 0.02 C are reported in the literature. Such low discharge rates limit the widespread use of metal fluoride active materials.
Another challenge for certain metal fluoride active materials is a significant hysteresis observed between the charge and discharge voltages during cycling. This hysteresis is typically on the order of about 1.0V to about 1.5V. While the origin of this hysteresis is uncertain, current evidence suggests that kinetic limitations imposed by low conductivity play an important role. Further, asymmetry in the reaction paths upon charge and discharge may also play a role. Since the electrochemical potential for many of the metal fluorides is on the order of 3.0V, this hysteresis of about 1.0V to about 1.5V limits the overall energy efficiency to approximately 50%.
Limited cycle life is another challenge for certain metal fluoride active materials. Although rechargeability has been demonstrated for many metal fluoride active materials, their cycle life is typically limited to tens of cycles and is also subject to rapid capacity fade. Three mechanisms are currently believed to limit the cycle life for the metal fluoride active materials: agglomeration of metallic nanoparticles and mechanical stress due to volume expansion. It is believed that metal fluoride active materials can cycle by virtue of the formation during lithiation of a continuous metallic network within a matrix of insulating LiF. As the number of cycles increases, the metal particles tend to accumulate together into larger, discrete particles. The larger agglomerated particles in turn create islands that are electrically disconnected from one another, thus reducing the capacity and ability to cycle the metal fluoride active materials. The second limitation to extended cycle life is the mechanical stress imparted to the binder materials by the metal fluoride particles as a result of the volume expansion that occurs during the conversion reaction. Over time, the binder is pulverized, compromising the integrity of the cathode. Notably, for the metal fluoride CuF2, no demonstrations of rechargeability have been reported.
For CuF2, an additional challenge prevents rechargeability. The potential required to recharge a CuF2electrode is 3.55V. However, in typical electrolytes for lithium ion batteries, Cu metal oxidizes to Cu2+ at approximately 3.5 V vs. Li/Li+. The oxidized copper can migrate to the anode, where it is irreversibly reduced back to Cu metal. As a result, Cu dissolution competes with the recharge of Cu+2LiF to CuF2, preventing cycling of the cell. The Cu metal accumulating on the anode surface can increase the impedance and/or destroy the solid-electrolyte interphase (SEI) on the anode.
The following papers and patents are among the published literature on metal fluorides that employ mixed conductors that are not electrochemically active within the voltage window of the metal fluoride: Badway, F. et al., Chem. Mater., 2007, 19, 4129; Badway, F. et al., J. Electrochem. Soc., 2007, 150, A1318; “Bismuth fluoride based nanocomposites as electrode materials” U.S. Pat. No. 7,947,392; “Metal Fluoride And Phosphate Nanocomposites As Electrode Materials” US 2008/0199772; “Copper fluoride based nanocomposites as electrode materials” US 2006/0019163; and “Bismuth oxyfluoride based nanocomposites as electrode materials” U.S. Pat. No. 8,039,149.
Certain embodiments of the present invention can be used to form electrochemical cells having metal fluoride active material that exhibit improved rate performance, improved energy efficiency, and improved cycle life when compared to prior batteries. Thus, these and other challenges can be addressed by embodiments of the present invention described below.